Overview of methods for variant calling from next-generation sequence data

Overview of methods for variant calling from next-
generation sequence data

Thomas Keane,
Vertebrate Resequencing Informatics,
Wellcome Trust Sanger Institute
Email: tk2@sanger.ac.uk

Vertebrate Resequencing Informatics 22nd July, 2010

SAM/BAM Format

Proliferation of alignment formats over the years: Cigar, psl, gff, xml etc.
SAM (Sequence Alignment/Map) format
  Single unified format for storing read alignments to a reference genome
BAM (Binary Alignment/Map) format
  Binary equivalent of SAM
  Developed for fast processing/indexing
Advantages
  Can store alignments from most aligners
  Supports multiple sequencing technologies
  Supports indexing for quick retrieval/viewing
  Compact size (e.g. 112Gbp Illumina = 116Gbytes disk space)
  Reads can be grouped into logical groups e.g. lanes, libraries, individuals/
genotypes
  Supports second best base call/quality for hard to call bases
Possibility of storing raw sequencing data in BAM as replacement to SRF & fastq


Read Entries in SAM

No. Name Description
1 QNAME Query NAME of the read or the read pair
2 FLAG Bitwise FLAG (pairing, strand, mate strand, etc.)
3 RNAME Reference sequence NAME
4 POS 1-Based leftmost POSition of clipped alignment
5 MAPQ MAPping Quality (Phred-scaled)
6 CIGAR Extended CIGAR string (operations: MIDNSHP)
7 MRNM Mate Reference NaMe (‘=’ if same as RNAME)
8 MPOS 1-Based leftmost Mate POSition
9 ISIZE Inferred Insert SIZE
10 SEQ Query SEQuence on the same strand as the reference
11 QUAL Query QUALity (ASCII-33=Phred base quality)

Heng Li , Bob Handsaker , Alec Wysoker , Tim Fennell , Jue Ruan , Nils Homer , Gabor Marth , Goncalo Abecasis ,
Richard Durbin , and 1000 Genome Project Data Processing Subgroup (2009) The Sequence Alignment/Map
format and SAMtools, Bioinformatics, 25:2078-2079


Extended Cigar Format

Cigar has been traditionally used as a compact way to represent a
sequence alignment
Operations include
  M - match or mismatch
  I - insertion
  D - deletion
SAM extends these to include
  S - soft clip
  H - hard clip
  N - skipped bases
  P – padding
E.g. Read: ACGCA-TGCAGTtagacgt

Ref:
ACTCAGTG—-GT

Cigar: 5M1D2M2I2M7S


What is the cigar line?

E.g. Read: tgtcgtcACGCATG---CAGTtagacgt

Ref:
ACGCATGCGGCAGT

Cigar:


Read Group Tag

Each lane has a unique RG tag
1000 Genomes
  Meta information derived from DCC
RG tags
  ID: SRR/ERR number
  PL: Sequencing platform
  PU: Run name
  LB: Library name
  PI: Insert fragment size
  SM: Individual
  CN: Sequencing center


1000 Genomes BAM File

samtools view –h mybam.bam

SAM/BAM Tools

Well defined specification for SAM/BAM
Several tools and programming APIs for interacting with SAM/BAM files
  Samtools - Sanger/C (http://samtools.sourceforge.net)
  Convert SAM <-> BAM
  Sort, index, BAM files
  Flagstat – summary of the mapping flags
  Merge multiple BAM files
  Rmdup – remove PCR duplicates from the library preparation
  Picard - Broad Institute/Java (http://picard.sourceforge.net)
  MarkDuplicates, CollectAlignmentSummaryMetrics, CreateSequenceDictionary,
SamToFastq, MeanQualityByCycle, FixMateInformation…….
  Bio-SamTool – Perl (http://search.cpan.org/~lds/Bio-SamTools/)
  Pysam – Python (http://code.google.com/p/pysam/)
BAM Visualisation
  BamView, LookSeq, Gap5: http://www.sanger.ac.uk/Software
  IGV: http://www.broadinstitute.org/igv/v1.3
  Tablet: http://bioinf.scri.ac.uk/tablet/


BAM Visualisation

http://www.sanger.ac.uk/mousegenomes

SNP Calling

SNP – single nucleotide polymorphisms
View the bases over a reference position and look for differences
Homozygous vs heterozygous SNPs
Factors to consider when calling SNPs
  Base call qualities of each supporting base
  Proximity to
 Small indel
 Homopolymer run (>4-5bp for 454 and >10bp for illumina)
  Mapping qualities of the reads supporting the SNP
 Low mapping qualities indicates repetitive sequence
  Read length
 Possible to align reads with high confidence to larger portion of the genome with
longer reads
  Paired reads
  Sequencing depth
 Few individuals/strains at high coverage vs. low coverage many individuals/strains
  1000 genomes is low coverage sequencing across many individuals
  Population based SNP calling methods


Read Length


Mouse SNP


Is this a SNP?


Short indel Calling

Small insertions and deletions observed in the alignment of the read
relative to the reference genome
BAM format
  I or D character denote indel in the read
Simple method
  Call indels based on the I or D events in the BAM file
 Samtools varFilter
Factors to consider when calling indels
  Misalignment of the read
 Alignment scoring - often cheaper to introduce multiple SNPs than an indel
 Sufficient flanking sequence either side of the read
  Homopolymer runs either side of the indel
  Length of the reads
  Homozygous or heterozygous


Example Indel


Is this an indel?


Local Realignment

Simple models for calling indels based on the initial alignments show
high false positives and negatives e.g samtools
More sophisticated algorithms currently being developed
  E.g. Dindel, GATK
Example Algorithm overview
  Scan for all I or D operations across the input BAM file
  Foreach I or D operation
 Create new haplotype based on the indel event
 Realign the reads onto the alternative reference
 Count the number of reads that support the indel in the alternative reference
 Make the indel call
Issues
  Very computationally intensive if testing every possible indel
 Alternatively test a subset of known indels (i.e. genotyping mode)


Structural Variation

Several types of structural variations (SVs)
  Large Insertions/deletions 76bp 76bp
  Inversions
300bp
  Translocations
  Copy number variations
Read pair information used to detect these events
  Paired end sequencing of either end of DNA
fragment
 Observe deviations from the expected fragment size
  Presence/absence of mate pairs
  Read depth to detect copy number variations
  Several SV callers published recently
Run several callers and produce large set of
partially overlapping calls


SV Types


Structural Variations

Medvedev et al, Nat Meth, 6(11), 2009


What is this?


What is this?

Mate pairs align in the same orientation


Tomorrow’s Lab 11-12

BAM Files
  Using samtools to manipulate BAM files
  Visualising reads in a BAM file
SNP Calling
  Calling SNPs from a BAM file
Variant Call Format (VCF)
  Introduction to VCF for storing SNPs and meta information
VCFTools
  Manipulating/comparing/intersecting lists of SNPs in VCF format


Overview of methods for variant calling from next-generation sequence data

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (19)

En vedette

En vedette (20)

Similaire à Overview of methods for variant calling from next-generation sequence data

Similaire à Overview of methods for variant calling from next-generation sequence data (20)

Plus de Thomas Keane

Plus de Thomas Keane (7)

Overview of methods for variant calling from next-generation sequence data